BRS10a quick overview
The systolic pressure (SP) input series is generated by an APR31a module (arterial blood pressure analyzer). BRS10a copies the SP input series at the Systolic pressure output (with or without filtering), calculates the pulse interval (PI) between the current SP and the following one for each SP, and outputs the PI at the current SP time point.
The user can exclude outliers from both SP and PI signals using absolute thresholds available in the module properties window (Figure 1).
Sequences of systolic pressures and pulse intervals are identified in the blood pressure signal and a linear regression is performed on each sequence to produce the Linear regression: Slope output and the associated correlation coefficient output (Linear regression: R).
The type of sequence (ascending or descending) is indicated by the Sequence output. The Sequence, Slope and R time values are placed at the first SP time point of each detected sequence.
The sequences are counted (N) on contiguous analysis windows with a length configurable in BRS10a properties window. Complementary parameters for baroreflex analysis are also calculated based on the analysis windows, like baroreflex effectiveness index (BEI). For more information about the calculated parameters see BRS10a reference document.
The analysis can be performed on selected windows to avoid zones with artifact, noise, signal loss, etc. The zone selection can be done in NOTOCORD-hem™, or in Microsoft Excel®, and is described in the application note How to extract baroreflex sequences on different analysis windows.
The applicability of the sequence technique to the estimation of spontaneous baroreflex sensitivity to resting mice is shown in . The analysis is based on a number of 40 windows with duration of 51.2 s, duration selected to facilitate the comparison with the spectral method in the cited paper. The spectral method can be applied using our cross-spectrum analyzer, CSA10a.
For our use case, we processed a recording of 2000 s of mouse pressure signal, detected the SP with APR31a and analyzed the SP series with BRS10a, based on the default parameters: analysis window = 50 s, synchronization delay (between SP ramp and PI ramp) = 0 cycles, minimum sequence length = 3 cycles.
A mean of 26.2 sequences were detected per window and a mean value of 3.59 ms/mmHg was obtained for the slope, as an index of the baroreflex sensitivity.
The reference  shows a further discrimination of the sequences by calculating the first order differences of SP and PI (Figure 2) and imposing a threshold for SP differences (thSP = 0, 0.2, 0.4 … mmHg) and a threshold for PI differences (thPI = 0, 1, 2 … ms). A sequence is valid if the first order differences are higher than the defined threshold.
The authors conclude that this discrimination seems not mandatory when spontaneous baroreflex sequences are investigated in mice, however it could be useful in other experimental conditions or species, and is easily implemented in Microsoft Excel®.
This additional discrimination is shown in Figure 3. SP series is extracted in column N, PI series is extracted in column O and the first order differences are calculated in columns P and Q respectively.
A sequence that passes the threshold criteria #1 to #3 receives a 1 in the corresponding cell of columns F to H. The “1” are used to count the sequences and to calculate a conditional average for the baroreflex sensitivity.
The formula used to discriminate the sequences using SP and PI thresholds are shown in Table 1.
As expected, higher thresholds in SP and PI dramatically reduce the number of sequences, as shown in .
 Dominique Laude, Véronique Baudrie and Jean-Luc Elghozi, Applicability of recent methods used to estimate spontaneous baroreflex sensitivity to resting mice. American journal of physiology. Regulatory, integrative and comparative physiology 294, R142-R150, 2008.